Merge tag 'upstream/2015.11.10'

[youtubedl] / youtube_dl / extractor / zdf.py

diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py
index 9ff00e26c4235e0eaace73b219b945ea17cee175..a795f56b37bbd710c895b0e255342057a5aa354f 100644 (file)
--- a/youtube_dl/extractor/zdf.py
+++ b/youtube_dl/extractor/zdf.py
@@ -1,12 +1,15 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
 # coding: utf-8
 from __future__ import unicode_literals
 

+import functools

 import re
 
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
     unified_strdate,
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
     unified_strdate,

+    OnDemandPagedList,
+    xpath_text,

 )
 
 
 )
 
 


@@ -17,13 +20,11 @@ def extract_from_xml_url(ie, video_id, xml_url):
         errnote='Failed to download video info')
 
     title = doc.find('.//information/title').text
         errnote='Failed to download video info')
 
     title = doc.find('.//information/title').text

-    description = doc.find('.//information/detail').text
-    duration = int(doc.find('.//details/lengthSec').text)
-    uploader_node = doc.find('.//details/originChannelTitle')
-    uploader = None if uploader_node is None else uploader_node.text
-    uploader_id_node = doc.find('.//details/originChannelId')
-    uploader_id = None if uploader_id_node is None else uploader_id_node.text
-    upload_date = unified_strdate(doc.find('.//details/airtime').text)
+    description = xpath_text(doc, './/information/detail', 'description')
+    duration = int_or_none(xpath_text(doc, './/details/lengthSec', 'duration'))
+    uploader = xpath_text(doc, './/details/originChannelTitle', 'uploader')
+    uploader_id = xpath_text(doc, './/details/originChannelId', 'uploader id')
+    upload_date = unified_strdate(xpath_text(doc, './/details/airtime', 'upload date'))

 
     def xml_to_format(fnode):
         video_url = fnode.find('url').text
 
     def xml_to_format(fnode):
         video_url = fnode.find('url').text


@@ -38,15 +39,14 @@ def extract_from_xml_url(ie, video_id, xml_url):
         ext = format_m.group('container')
         proto = format_m.group('proto').lower()
 
         ext = format_m.group('container')
         proto = format_m.group('proto').lower()
 

-        quality = fnode.find('./quality').text
-        abr = int(fnode.find('./audioBitrate').text) // 1000
-        vbr_node = fnode.find('./videoBitrate')
-        vbr = None if vbr_node is None else int(vbr_node.text) // 1000
+        quality = xpath_text(fnode, './quality', 'quality')
+        abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000)
+        vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000)

 
 

-        width_node = fnode.find('./width')
-        width = None if width_node is None else int_or_none(width_node.text)
-        height_node = fnode.find('./height')
-        height = None if height_node is None else int_or_none(height_node.text)
+        width = int_or_none(xpath_text(fnode, './width', 'width'))
+        height = int_or_none(xpath_text(fnode, './height', 'height'))
+
+        filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize'))

 
         format_note = ''
         if not format_note:
 
         format_note = ''
         if not format_note:


@@ -62,12 +62,31 @@ def extract_from_xml_url(ie, video_id, xml_url):
             'vbr': vbr,
             'width': width,
             'height': height,
             'vbr': vbr,
             'width': width,
             'height': height,

-            'filesize': int_or_none(fnode.find('./filesize').text),
+            'filesize': filesize,

             'format_note': format_note,
             'protocol': proto,
             '_available': is_available,
         }
 
             'format_note': format_note,
             'protocol': proto,
             '_available': is_available,
         }
 

+    def xml_to_thumbnails(fnode):
+        thumbnails = []
+        for node in fnode:
+            thumbnail_url = node.text
+            if not thumbnail_url:
+                continue
+            thumbnail = {
+                'url': thumbnail_url,
+            }
+            if 'key' in node.attrib:
+                m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key'])
+                if m:
+                    thumbnail['width'] = int(m.group(1))
+                    thumbnail['height'] = int(m.group(2))
+            thumbnails.append(thumbnail)
+        return thumbnails
+
+    thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage'))
+

     format_nodes = doc.findall('.//formitaeten/formitaet')
     formats = list(filter(
         lambda f: f['_available'],
     format_nodes = doc.findall('.//formitaeten/formitaet')
     formats = list(filter(
         lambda f: f['_available'],


@@ -79,6 +98,7 @@ def extract_from_xml_url(ie, video_id, xml_url):
         'title': title,
         'description': description,
         'duration': duration,
         'title': title,
         'description': description,
         'duration': duration,

+        'thumbnails': thumbnails,

         'uploader': uploader,
         'uploader_id': uploader_id,
         'upload_date': upload_date,
         'uploader': uploader,
         'uploader_id': uploader_id,
         'upload_date': upload_date,


@@ -87,7 +107,7 @@ def extract_from_xml_url(ie, video_id, xml_url):
 
 
 class ZDFIE(InfoExtractor):
 
 
 class ZDFIE(InfoExtractor):

-    _VALID_URL = r'^https?://www\.zdf\.de/ZDFmediathek(?P<hash>#)?/(.*beitrag/(?:video/)?)(?P<id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
+    _VALID_URL = r'(?:zdf:|zdf:video:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/(.*beitrag/(?:video/)?))(?P<id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'

 
     _TEST = {
         'url': 'http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt',
 
     _TEST = {
         'url': 'http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt',


@@ -106,6 +126,52 @@ class ZDFIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

-

         xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
         return extract_from_xml_url(self, video_id, xml_url)
         xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
         return extract_from_xml_url(self, video_id, xml_url)

+
+
+class ZDFChannelIE(InfoExtractor):
+    _VALID_URL = r'(?:zdf:topic:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/.*kanaluebersicht/)(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://www.zdf.de/ZDFmediathek#/kanaluebersicht/1586442/sendung/Titanic',
+        'info_dict': {
+            'id': '1586442',
+        },
+        'playlist_count': 3,
+    }
+    _PAGE_SIZE = 50
+
+    def _fetch_page(self, channel_id, page):
+        offset = page * self._PAGE_SIZE
+        xml_url = (
+            'http://www.zdf.de/ZDFmediathek/xmlservice/web/aktuellste?ak=web&offset=%d&maxLength=%d&id=%s'
+            % (offset, self._PAGE_SIZE, channel_id))
+        doc = self._download_xml(
+            xml_url, channel_id,
+            note='Downloading channel info',
+            errnote='Failed to download channel info')
+
+        title = doc.find('.//information/title').text
+        description = doc.find('.//information/detail').text
+        for asset in doc.findall('.//teasers/teaser'):
+            a_type = asset.find('./type').text
+            a_id = asset.find('./details/assetId').text
+            if a_type not in ('video', 'topic'):
+                continue
+            yield {
+                '_type': 'url',
+                'playlist_title': title,
+                'playlist_description': description,
+                'url': 'zdf:%s:%s' % (a_type, a_id),
+            }
+
+    def _real_extract(self, url):
+        channel_id = self._match_id(url)
+        entries = OnDemandPagedList(
+            functools.partial(self._fetch_page, channel_id), self._PAGE_SIZE)
+
+        return {
+            '_type': 'playlist',
+            'id': channel_id,
+            'entries': entries,
+        }